Popular character forms (Suzi) and semantic compound (Huiyi) characters in medieval Chinese manuscripts.

Author:Galambos, Imre

Traditional Chinese scholarship visualized orthographic structure using the system of liush[u.bar] [TEXT NOT REPRODUCIBLE IN ASCII], a technical term explained as six principles of character formation. According to extant sources these categories were first put forth towards the end of the first century C.E. and have been in use ever since. Although individual categories had been sporadically criticized as early as the Song period, it was during the philological renaissance of the Qing dynasty when it became clear that the entire classification system had to be reconsidered. Later on, the discovery of oracle-bone inscriptions and pre-Han manuscripts provided additional material for reassessing the claims of traditional historiography on the evolution of Chinese writing and the forces behind it. Among the problematic categories within the Naha system was the principle of huiyi [TEXT NOT REPRODUCIBLE IN ASCII], at times rendered into English as 'syssemantic characters' or simply 'semantic compounds'. (1) This traditional category proved to be often based on folk etymologies, and most examples of it can be demonstrated to contain a phonetic component, in contrast with the traditional view that saw them as purely semantic combinations.

While modern research may be justified in doubting the impact of the huiyi principle at the early stages of the writing system, one cannot fail to notice the presence of numerous hidyi-type forms in medieval manuscripts and epigraphic sources. Some of these forms commonly feature in medieval dictionaries, while others are seen only in manuscripts and inscriptions. To be sure, for the most part these are variant forms of characters with otherwise well-attested phonetic origins, yet their occurrence in the post-Han period is a phenomenon that deserves our attention. In this paper, I propose to look at some of the popular or non-standard forms (suzi [TEXT NOT REPRODUCIBLE IN ASCII]) found in medieval manuscripts and dictionaries in an attempt to reconsider the huiyi category from the point of view of the manuscript tradition. Rather than discussing the etymology and early development of established characters, I am specifically interested in non-standard character forms used in everyday writing, because these demonstrate that even if the huiyi principle did not play a major role during the early stages of the Chinese script, by medieval times it was certainly one of the key models according to which people understood orthographic structure.


Traditional Chinese scholarship described the principles behind the evolution of characters in terms of the liush[u.bar] ('six scripts' or 'six types of writing'). Although this term appears in the Zh[o.bar]uli [TEXT NOT REPRODUCIBLE IN ASCII], its use with respect to character structure dates to the first century C.E. when it surfaced in three different sources. The most elaborate of these is the Shu[o.bar]wen jiezi [TEXT NOT REPRODUCIBLE IN ASCII] by Xu Shen [TEXT NOT REPRODUCIBLE IN ASCII](ca. 58-ca. 147), who explained the six categories in the "Post-face" [TEXT NOT REPRODUCIBLE IN ASCII] in the following way:

[TEXT NOT REPRODUCIBLE IN ASCII] According to the Zh[o.bar]uli, school begins at the age of eight. When the Protector teaches the sons of the state, he begins with the liush[u.bar]. The first of these is zhishi ('pointing at things') (2) Zhishi characters are the ones that can be understood by looking at them, the meaning of which can be seen through observation. The characters [TEXT NOT REPRODUCIBLE IN ASCII] and [TEXT NOT REPRODUCIBLE IN ASCII] are like this. The second is xiangxing ('depicting form'). Xiangxing characters are the ones that depict objects by reproducing their physical shape. The characters [TEXT NOT REPRODUCIBLE IN ASCII] and [TEXT NOT REPRODUCIBLE IN ASCII] are like this. The third is xingsh[e.bar]ng ('form and sound'). Xingsh[e.bar]ng characters are the ones that take a thing/object to indicate the name and combine it with a [phonetic] semblance. The characters [TEXT NOT REPRODUCIBLE IN ASCII] and [TEXT NOT REPRODUCIBLE IN ASCII] are like this. The fourth is huiyi ('joining ideas').huiyi characters are the ones that conjoin categories to present the indicated meaning. The characters [TEXT NOT REPRODUCIBLE IN ASCII] and [TEXT NOT REPRODUCIBLE IN ASCII] are like this. The fifth is zhuanzhu ('commenting by rotation').Zhuanzhu characters are the ones that establish categories based on a single origin and that borrow their analogous meanings from each other. The characters [TEXT NOT REPRODUCIBLE IN ASCII] and [TEXT NOT REPRODUCIBLE IN ASCII] are like this. The sixth is jiajie ('borrowed'). jiajie characters are the ones that are assigned a written from, which did not exist originally, based on their pronunciation. The characters [TEXT NOT REPRODUCIBLE IN ASCII] and [TEXT NOT REPRODUCIBLE IN ASCII] are like this. In addition to listing the names of the six categories, Xiii Shen provides two examples for each. He also gives a short gloss of each term, and this is the only evidence we have today of how the categories might have been understood in Han times. (3) The other Han source describing the six principles is the "Yiwenzhi."[TEXT NOT REPRODUCIBLE IN ASCII] (Record of Arts and Letters) chapter in B[a.bar]n Gu' s [TEXT NOT REPRODUCIBLE IN ASCII] (32-92 C.E.) Hansh[u.bar] [TEXT NOT REPRODUCIBLE IN ASCII] 'in (History of the Him Dynasty), which retells the history of Chinese writing in very much the same manner as Xu Shen's "Postface." Indeed, the two accounts show a number of similarities that confirm that they ultimately go back to the same source. (4) At the same time, B[a.bar]n Gu's account of the liush[u.bar] is more concise, only giving a list of names of the categories without examples: xiangxing [TEXT NOT REPRODUCIBLE IN ASCII], xiangshi [TEXT NOT REPRODUCIBLE IN ASCII], xiangyi [TEXT NOT REPRODUCIBLE IN ASCII], xiangsh[e.bar]ng [TEXT NOT REPRODUCIBLE IN ASCII], zhuanzhu [TEXT NOT REPRODUCIBLE IN ASCII], and jiajie [TEXT NOT REPRODUCIBLE IN ASCII]. In addition, we learn that these represent "the basis of character formation" [TEXT NOT REPRODUCIBLE IN ASCII]. (5)( B.C.E.-83 c.E.) commentary to the Zh[o.bar]uli from the second half of the first century C.E., identified the six categories as xiangxing [TEXT NOT REPRODUCIBLE IN ASCII], huiyi [TEXT NOT REPRODUCIBLE IN ASCII], zhuanzhu [TEXT NOT REPRODUCIBLE IN ASCII], chushi [TEXT NOT REPRODUCIBLE IN ASCII], jiajie [TEXT NOT REPRODUCIBLE IN ASCII], and xiesh[e.bar]ng [TEXT NOT REPRODUCIBLE IN ASCII], which yet again shows some discrepancy with both Xu Shen's and Ban Gu's terminology. (6) Thus at the source of the tradition we have three different authors from the Eastern Han period with three sets of names, most likely describing the same or very similar principles using slightly different nomenclature. (7) In later centuries, however, the liush[u.bar] system developed into a complex conceptual framework that formed the theoretical foundation of the branch of Chinese philology dealing with the graphic shape of characters. Indeed, since the Song period this branch was often referred to by the name of liush[u.bar]xue [TEXT NOT REPRODUCIBLE IN ASCII], or the "study of the six principles of writing." (8)

But early on, critics of individual categories raised objections regarding the etymological correctness of these principles. In the Song dynasty, for example, Zheng Qiao [TEXT NOT REPRODUCIBLE IN ASCII] (1104-62) noted in his Liush[u.bar] lue [TEXT NOT REPRODUCIBLE IN ASCII] (Outline of the Six Scripts) that in the body of the Shu[o.bar]wen Xu Shen effectively only used the xiangxing and xingsh[e.bar]ng categories. During the Qing dynasty, along with a renewed interest in epigraphy and textual studies, the Shu[o.bar]wen and the liush[u.bar] once again became the focus of scholarly investigation and reinterpretation. (9) Finally in modem times the rapidly growing number of newly discovered inscriptions and early manuscripts provided abundant firsthand material for creating a historically plausible narrative of the development of the Chinese script. Especially the Shang oracle-bone inscriptions, which to this day remain the earliest surviving examples of writing in China, have come to play an important role in understanding the earlier stages of the history of the script. But equally important were the bronze inscriptions and bamboo-slip documents, not only because they provided fresh sources for research but also because these spectacular discoveries gave an impetus that drove the entire field forward.

With the adoption of Western linguistic theories for the study of the Chinese writing system, an increasing amount of attention has been directed to the phonetic nature of the script, and these new findings were also applied to explaining character etymology. As part of this new approach, the traditional categories of liush[u.bar], and especially the two distinctly non-phonetic principles of zhuanzhu and huiyi, came under suspicion. While what zhuanzhu entailed remains a subject of scholarly debate, (10) the meaning of the huiyi category is fairly unambiguous: it is a principle according to which two or more components are joined into a single character and their semantic values together come to represent the semantic value of the new composite character. As examples in the "Postface" to the Shu[o.bar]wen, Xu Shen gave the characters [TEXT NOT REPRODUCIBLE IN ASCII] and [TEXT NOT REPRODUCIBLE IN ASCII], of which at least the former had a literary precedent, having been explained in the Zuozhuan [TEXT NOT REPRODUCIBLE IN ASCII] in terms of its orthography: "the meaning of `martial' refers to halting weapons" [TEXT NOT REPRODUCIBLE IN ASCII]. Since in the Shu[o.bar]wen we find only three more characters explicitly identified as huiyi compounds (i.e., [TEXT NOT REPRODUCIBLE IN ASCII], [TEXT NOT REPRODUCIBLE IN ASCII]...

To continue reading